DeepSeek VL2 Tiny
Multimodal
Zero-eval
#3MME
by DeepSeek
About
DeepSeek VL2 Tiny is a multimodal language model developed by DeepSeek. It achieves strong performance with an average score of 63.1% across 14 benchmarks. It excels particularly in DocVQA (88.9%), ChartQA (81.0%), OCRBench (80.9%). As a multimodal model, it can process and understand text, images, and other input formats seamlessly. Released in 2024, it represents DeepSeek's latest advancement in AI technology.
Timeline
AnnouncedDec 13, 2024
ReleasedDec 13, 2024
Specifications
Capabilities
Multimodal
License & Family
License
deepseek
Benchmark Performance Overview
Performance metrics and category breakdown
Overall Performance
14 benchmarks
Average Score
63.1%
Best Score
88.9%
High Performers (80%+)
4Top Categories
vision
69.1%
general
62.5%
math
53.6%
roleplay
53.2%
Benchmark Performance
Top benchmark scores with normalized values (0-100%)
Ranking Across Benchmarks
Position relative to other models on each benchmark
DocVQA
Rank #20 of 26
#17Llama 3.2 90B Instruct
90.1%
#18Pixtral-12B
90.7%
#19DeepSeek VL2 Small
92.3%
#20DeepSeek VL2 Tiny
88.9%
#21Llama 3.2 11B Instruct
88.4%
#22Gemma 3 12B
87.1%
#23Gemma 3 27B
86.6%
ChartQA
Rank #20 of 24
#17Phi-4-multimodal-instruct
81.4%
#18Phi-3.5-vision-instruct
81.8%
#19Pixtral-12B
81.8%
#20DeepSeek VL2 Tiny
81.0%
#21Gemma 3 27B
78.0%
#22Grok-1.5V
76.1%
#23Gemma 3 12B
75.7%
OCRBench
Rank #7 of 7
#4DeepSeek VL2
81.1%
#5DeepSeek VL2 Small
83.4%
#6Phi-4-multimodal-instruct
84.4%
#7DeepSeek VL2 Tiny
80.9%
TextVQA
Rank #7 of 15
#4Nova Pro
81.5%
#5DeepSeek VL2 Small
83.4%
#6DeepSeek VL2
84.2%
#7DeepSeek VL2 Tiny
80.7%
#8Nova Lite
80.2%
#9Grok-1.5V
78.1%
#10Phi-4-multimodal-instruct
75.6%
AI2D
Rank #17 of 17
#14Gemma 3 4B
74.8%
#15Phi-3.5-vision-instruct
78.1%
#16DeepSeek VL2 Small
80.0%
#17DeepSeek VL2 Tiny
71.6%
All Benchmark Results for DeepSeek VL2 Tiny
Complete list of benchmark scores with detailed information
DocVQA DocVQA benchmark | vision | multimodal | 0.89 | 88.9% | Self-reported |
ChartQA ChartQA benchmark | general | multimodal | 0.81 | 81.0% | Self-reported |
OCRBench OCRBench benchmark | general | text | 0.81 | 80.9% | Self-reported |
TextVQA TextVQA benchmark | vision | multimodal | 0.81 | 80.7% | Self-reported |
AI2D AI2D benchmark | general | text | 0.72 | 71.6% | Self-reported |
MMBench MMBench benchmark | general | text | 0.69 | 69.2% | Self-reported |
MMBench-V1.1 MMBench-V1.1 benchmark | general | text | 0.68 | 68.3% | Self-reported |
InfoVQA InfoVQA benchmark | vision | multimodal | 0.66 | 66.1% | Self-reported |
RealWorldQA RealWorldQA benchmark | general | text | 0.64 | 64.2% | Self-reported |
MathVista MathVista benchmark | math | text | 0.54 | 53.6% | Self-reported |
Showing 1 to 10 of 14 benchmarks